Abatacept in rheumatoid arthritis: survival on drug, clinical outcomes, and their predictors—data from a large national quality register

BACKGROUND: There are limited data regarding efficacy of abatacept treatment for rheumatoid arthritis (RA) outside clinical trials. Quality registers have been useful for observational studies on tumor necrosis factor inhibition in clinical practice. The aim of this study was to investigate clinical efficacy and tolerability of abatacept in RA, using a national register. METHODS: RA patients that started abatacept between 2006 and 2017 and were included in the Swedish Rheumatology Quality register (N = 2716) were investigated. Survival on drug was estimated using Kaplan-Meier analysis. The European League Against Rheumatism (EULAR) good response and Health Assessment Questionnaire (HAQ) response (improvement of ≥ 0.3) rates (LUNDEX corrected for drug survival) at 6 and at 12 months were assessed. Predictors of discontinuation were investigated by Cox regression analyses, and predictors of clinical response by logistic regression. Significance-based backward stepwise selection of variables was used for the final multivariate models. RESULTS: There was a significant difference in drug survival by previous biologic disease-modifying antirheumatic drug (bDMARD) exposure (p < 0.001), with longer survival in bionaïve patients. Men (hazard ratio (HR) 0.86, 95% confidence interval (CI) 0.74–0.98) and methotrexate users (HR 0.85, 95% CI 0.76–0.95) were less likely to discontinue abatacept, whereas a high pain score predicted discontinuation (HR 1.14 per standard deviation, 95% CI 1.07–1.20). The absence of previous bDMARD exposure, male sex, and a low HAQ score were independently associated with LUNDEX-corrected EULAR good response. The absence of previous bDMARD exposure also predicted LUNDEX-corrected HAQ response. CONCLUSIONS: In this population-based study of RA, bDMARD naïve patients and male patients were more likely to remain on abatacept with a major clinical response.